Raymond's Weblog^


I think I went through all the classic stages of being a startup guy in about two years of university

The thing itself began as an adaptation of the Stanford Marriage Pact for Oxford, and later Edinburgh and Leeds. We focused on asking the kinds of questions that would

Th algorithm had two parts: one to evaluate how good a match two people were, and another which used those scores to give every user one or two specific matches. Then we picked out some specific prompt questions and put them into emails that got fired out to all our users at the stroke of midnight, give or take. One nice feature of this setup was that we could work manically for a couple nights and then go back to our degrees for the next month or two.

How it feels to build

The main takeaway by far was how it feels to actually be building something. We spent a solid year planning out what things we might build, assembling a team of people who had meetings and discussed strategies. We had a logo and a facebook page and some freelance developers in theory making us a full stack web app.

All that turned out to be far less valuable than a typeform wired into a spreadsheet with a massive apps script. It helped that I’d read the same advice from every startup person ever, but the lesson doesn’t really sink in until you’re literally building something for real people: you just figure out what works and do it.

And you use all the tools you have. Our main thing was getting people to fill out a survey, to get matches, but we also used the survey to gather promotional material (fun graphics) and even look for collaborators (do you like graphic design? would you want to help us?).

Initially a lot of our exposure came from student-run facebook and instagram pages; we had the advantage that people trusted us because we were students. And because these things were maintained by people who were busy and/or graduating, we were able to sometimes step in and take them over, so that we could just have the distribution channels.

Value differences grow with power

The plan was to make a cute app, but reasonably quickly we were sitting on fairly sensitive information (the kind you need to match romantic partners) from literally thousands of people, and we were running some of the main places where students interacted online.

Oxford is an odd place. Some people dedicate most of their time to trying to become, say, president of the Oxford Union. And in many ways this is just a glorified popularity competition, but the winners often end up being things like Prime Minister. People fight dirty, and every year there’s some kind of scandal. Then there’s the student newspapers, the other political societies, and a thousand other people who are trying to work out what to do with their lives.

All I’ll say is, I knew my cofounders and I had slightly different opinions on things like censorship, freedom of expression, impartiality, and so on - we were there studying things like politics and philosophy, so it’d be kind of a shame if we didn’t. But suddenly they mattered, because our decisions kind of actually mattered. And I’m glad that I learned that in university.

Just going for the thing

My favourite memories are of hearing people talking about something I built as if it were a real institution. A friend of mine dated a girl for a few months, and only after they’d stopped seeing each other did he admit that they’d met through something I built. I expect to be counterfactually responsible for at least one human being.

At my first job, there was a poster of Steve Jobs on the wall saying “Everything around you that you call life was made up by people that were no smarter than you, and you can change it!”